Star Attention: Efficient LLM Inference over Long Sequences
Shantanu Acharya, Fei Jia, Boris Ginsburg
Abstract
Inference with Transformer-based Large Language Models (LLMs) on long sequences is both costly and slow due to the quadratic complexity of the self-attention mechanism. We introduce Star Attention, a two-phase block-sparse approximation that improves computational efficiency by sharding attention across multiple hosts while minimizing communication overhead. In the first phase, the context is processed using blockwise-local attention across hosts, in parallel. In the second phase, query and response tokens attend to all prior cached tokens through sequence-global attention. Star Attention integrates seamlessly with most Transformer-based LLMs trained with global attention, reducing memory requirements and inference time by up to 11x while preserving 97-100% of accuracy.
BibTeX
@inproceedings{
acharya2025star,
title={Star Attention: Efficient {LLM} Inference over Long Sequences},
author={Shantanu Acharya and Fei Jia and Boris Ginsburg},
booktitle={Forty-second International Conference on Machine Learning},
year={2025},
url={https://openreview.net/forum?id=QY7Au9nZwp}
}